Probing the Basis of Domain-Dependent Inhibition Using Novel

May 6, 2008 - domains, with differing substrate preferences. X-ray crystal structures of the C and N domains complexed with various inhibitors have al...
0 downloads 10 Views 3MB Size
Download PDF
5942

Biochemistry 2008, 47, 5942–5950

Probing the Basis of Domain-Dependent Inhibition Using Novel Ketone Inhibitors of Angiotensin-Converting Enzyme†,‡ Jean M. Watermeyer, Wendy L. Kro¨ger, Hester G. O’Neill, B. Trevor Sewell, and Edward D. Sturrock* DiVision of Medical Biochemistry, Institute of Infectious Disease and Molecular Medicine, UniVersity of Cape Town, ObserVatory 7925, South Africa ReceiVed February 14, 2008; ReVised Manuscript ReceiVed April 2, 2008

ABSTRACT:

Human angiotensin-converting enzyme (ACE) has two homologous domains, the N and C domains, with differing substrate preferences. X-ray crystal structures of the C and N domains complexed with various inhibitors have allowed identification of active site residues that might be important for the molecular basis of this selectivity. However, it is unclear to what extent the different residues contribute to substrate domain selectivity. Here, cocrystal structures of human testis ACE, equivalent to the C domain, have been determined with two novel C domain-selective ketomethylene inhibitors, (5S)-5-[(Nbenzoyl)amino]-4-oxo-6-phenylhexanoyl-L-tryptophan (kAW) and (5S)-5-[(N-benzoyl)amino]-4-oxo-6phenylhexanoyl-L-phenylalanine (kAF). The ketone groups of both inhibitors bind to the zinc ion as a hydrated geminal diolate, demonstrating the ability of the active site to catalyze the formation of the transition state. Moreover, active site residues involved in inhibitor binding have been mutated to their N domain counterparts, and the effect of the mutations on inhibitor binding has been determined. The C domain selectivity of these inhibitors was found to result from interactions between bulky hydrophobic side chain moieties and C domain-specific residues F391, V518, E376, and V380 (numbering of testis ACE). Mutation of these residues decreased the affinity for the inhibitors 4-20-fold. T282, V379, E403, D453, and S516 did not contribute individually to C domain-selective inhibitor binding. Further domainselective inhibitor design should focus on increasing both the affinity and selectivity of the side chain moieties.

The existence of two homologous active sites in one polypeptide chain is a rare phenomenon, thought to be caused by gene duplication leading to the development of new activities (1, 2). Human somatic angiotensin-converting enzyme (ACE)1 is one such case, having two 55% identical domains, N and C, each containing an active site with a zinc binding motif, HEMGH (3, 4). ACE is a key target for the treatment of hypertension because of its role in increasing blood pressure via the renin-angiotensin system (5, 6). Further examples of such two-domain enzymes are sucrase-isomaltase (SI) and lactase-phlorizin hydrolase (LPH), † This work was supported by the Wellcome Trust, U.K. (Senior International Research Fellowship 070060 to E.D.S.), the National Research Foundation, South Africa, the Stella and Paul Loewenstein Charitable and Educational Trust, the Ernst and Ethel Erikson Trust, and the University of Cape Town. ‡ The atomic coordinates and structure factors (entries 3BKK and 3BKL) have been deposited in the Protein Data Bank. * To whom correspondence should be addressed: Division of Medical Biochemistry, Wernher Beit Building North, University of Cape Town, Observatory 7925, South Africa. Telephone: +2721 406 6312. Fax: +2721 406 6470. E-mail: [email protected]. 1 Abbreviations: ACE, angiotensin-converting enzyme; SI, sucraseisomaltase; LPH, lactase-phlorizin hydrolase; kAP, keto-ACE; kAW, (5S)-5-[(N-benzoyl)amino]-4-oxo-6-phenylhexanoyl-L-tryptophan; kAF, (5S)-5-[(N-benzoyl)amino]-4-oxo-6-phenylhexanoyl-L-phenylalanine; tACE, testis angiotensin-converting enzyme; tACE-G13, minimally glycosylated tACE mutant; DMSO, dimethyl sulfoxide; HHL, hippurylhistidyl-leucine; z-FHL, z-phenyl-histidyl-leucine; PDB, Protein Data Bank; CPA, carboxypeptidase A; BOP, 5-amino(N-tert-butoxycarbonyl)-2-benzyl-4-oxo-6-phenylhexanoic acid.

both involved in sugar metabolism (7, 8). In these enzymes, the homologous domains, which are ∼40% identical, bind similar but distinct substrates. The domains of SI cleave related disaccharides sucrose and isomaltose, while LPH cleaves lactose and the glucoside phlorizin. In the case of LPH, mechanism-based domain-selective inhibitors were used to elucidate the activities of the two domains (9). Other examples of enzymes containing homologous active domains include adenylyl transferase of Escherichia coli (10), which catalyzes the adenylylation and deadenylylation of glutamine synthetase, and protein disulfide isomerase (11), which catalyzes disulfide bond oxidation, reduction, and isomerization. In all of these examples, a few active site substitutions correspond to significantly different, but functionally complementary, substrate specificities for the homologous domains. ACE differs from these enzymes in that its activity is not highly specific, with each domain being able to cleave a range of peptide sequences; however, the efficiency of cleavage varies between domains, and some substrates are domain-specific (12–14). The N and C domains of ACE also differ in chloride dependence (4, 15) and thermostability (16). These differences between the domains make it appealing to develop an antihypertensive ACE inhibitor therapy that selectively targets one domain. The C domain has been shown to produce the potent vasopressor angiotensin II from angiotensin I more efficiently than the N domain, in ViVo (17–19). Moreover, buildup of bradykinin, a substrate of both domains, has been implicated in a number of ACE

10.1021/bi8002605 CCC: $40.75  2008 American Chemical Society Published on Web 05/06/2008

Domain Selectivity of Ketone ACE Inhibitors

FIGURE 1: Chemical structures of kAP and its derivatives kAW and kAF, showing residue positions relative to the zinc-binding group (P2, P1, P1′, P2′).

inhibitor-related side effects (20–22). It has been suggested that a C domain-selective ACE inhibitor would reduce blood pressure while allowing the N domain to maintain physiological bradykinin levels, resulting in more effective treatment of hypertension with reduced side effects (14, 23). Keto-ACE (hereafter called kAP), an analogue of the tripeptide ACE inhibitor Bz-Phe-Gly-Pro, has been shown to be 26-34 times more selective for the C domain with a variety of substrates (24, 25). Using kAP as a platform and making C-terminal substitutions, our group has synthesized a number of novel compounds. Two of these, (5S)-5-[(Nbenzoyl)amino]-4-oxo-6-phenylhexanoyl-L-tryptophan (kAW) and (5S)-5-[(N-benzoyl)amino]-4-oxo-6-phenylhexanoyl-Lphenylalanine (kAF) (Figure 1), are highly C domainselective (26). In this study, we determine the crystal structures of human testis ACE (tACE), which is equivalent to the C domain of ACE, in complex with the novel domain-selective inhibitors kAW and kAF. In addition, we assess the effect of a number of active site mutations on inhibitor binding. This combined structural and kinetic approach allows the determination of the extent to which particular residues contribute to the C domain selectivity of these inhibitors. EXPERIMENTAL PROCEDURES tACE Constructs. A fully N-glycosylated tACE mutant, tACE∆36NJ, truncated after S625 and lacking the 36 O-glycosylated, N-terminal residues (27) was cloned into the BamHI and EcoRI restriction sites of pcDNA3.1(+) (Invitrogen) for expression purposes. A minimally glycosylated form of this mutant, tACE-G13 (28), was used for crystallization. Inhibitors. Inhibitors kAP, kAW, and kAF were synthesized previously (26, 29). Stocks were made by dissolving inhibitors in a small volume of DMSO and then diluting in water or buffer, for enzyme assays. Cloning and Mutagenesis. Site-directed mutagenesis was performed on an SphI-EcoRI fragment of the C domain

Biochemistry, Vol. 47, No. 22, 2008 5943 coding region in pGEM11zf(+) (Promega) as described previously (30), converting selected active site residues to their N domain counterparts. Oligonucleotides containing the mutation of interest were synthesized by Inqaba Biotech, and reagents used for site-directed mutagenesis were supplied by Promega. Colonies were screened with appropriate restriction enzymes, and positive clones were confirmed by nucleotide sequencing. C domain mutants were constructed in pcDNA3.1+ for protein expression. Heterologous Expression and Purification. tACE constructs were expressed in Chinese hamster ovary cells and purified by Sepharose-lisinopril affinity chromatography, as described previously (28, 31). ACE activity was detected using a fluorogenic assay with substrates hippuryl-histidylleucine (HHL, Sigma) or z-phenyl-histidyl-leucine (z-FHL, Sigma) (32). Pooled fractions were dialyzed against 2 L of 0.5 mM Hepes (pH 7.5) overnight. tACE-G13 was concentrated to 1-5 mg/mL for crystallization using an Amicon Ultra-4 spin column (Millipore). The purified protein’s molecular weight and purity were assessed using SDS-PAGE. Crystallization. Crystals were grown by vapor diffusion at 16 °C in hanging drops comprising 2 µL of protein (1.45 mg/mL) or protein/inhibitor solution mixed with 2 µL of reservoir buffer over 1 mL reservoirs covered with 500 µL of oil (3 parts paraffin, 2 parts silicon oil; Hampton) (33). Inhibitor-free tACE-G13 microcrystals grew within 2 weeks over reservoirs containing 10 mM sodium acetate (pH 4.7) (Merck), 15% PEG 4000 (Fluka), and 10 µM ZnSO4. Cocrystals of tACE-G13 with inhibitors were grown by streak-seeding these microcrystals into fresh drops containing inhibitor mixed with tACE-G13, over reservoirs containing a higher concentration of sodium acetate (30 mM) but otherwise identical to the reservoirs used for growing microcrystals. Diffracting crystals were grown within 2 weeks in drops containing 1 mM kAW with 1 mg/mL tACEG13 and 1.25 mM kAF with 1.4 mg/mL tACE-G13. Data Collection and Processing. Data were collected at 100 K from a single crystal per inhibitor at beamline BM14(UK) of the European Synchrotron Radiation Facility (Grenoble, France) using a wavelength of 1.033 Å. Data were processed using HKL2000 (34), and phasing was carried out with EPMR 2.5 (35) using protein atoms from the crystal structure of tACE-G13 (PDB entry 2IUL) as a model. Model building was carried out using O version 9.0.7 (36), against 2Fobs - Fcalc, Fobs - Fcalc, and composite omit maps (simulated annealing with 5% omission), calculated using CNS (37). Waters were added using the water_pick protocol of CNS (37), followed by visual inspection. Ligand molecules were built using PRODRG (38) and ARP/wARP (39). Maximum likelihood minimization and simulated annealing of selected residues were carried out using CNS (37). PROCHECK and SFCHECK from the CCP4 program suite (39) and Molprobity (40) were used for model validation. Hydrogen bonds and close contacts were identified using HBplus (41). Figures were generated using Pymol version 0.99 (DeLano Scientific, Palo Alto, CA). Substrate Hydrolysis. The hydrolysis of the fluorogenic peptide Abz-FRK(Dnp)P-OH (a kind gift from A. Carmona) was characterized in a continuous assay at 25 °C under initial rate conditions, by measuring fluorescence at a λex of 320 nm and a λem of 420 nm. A substrate concentration range of

5944

Biochemistry, Vol. 47, No. 22, 2008

0-12 µM was added to the reaction mixture of 50 mM Hepes buffer (pH 6.8) containing 200 mM NaCl, 10 µM ZnCl2, and an enzyme concentration of 0.2 nM. Kinetic constants were calculated using the direct linear plot method (42, 43). Inhibition Kinetics. ACE constructs (2 nM) were incubated with an appropriate concentration range of inhibitor in 50 mM Hepes buffer (pH 6.8) containing 200 mM NaCl and 10 µM ZnCl2, at ambient temperature for 60 min. Twenty microliters of the enzyme/inhibitor mixture was added to a reaction mixture of 50 mM Hepes buffer (pH 6.8) containing 200 mM NaCl, 10 µM ZnCl2, and 4 or 8 µM AbzFRK(Dnp)P-OH, to create a total volume of 300 µL, in triplicate. Residual enzyme activity was monitored by determining the fluorescence at a λex of 320 nm and a λem of 420 nm, at 25 °C. Inhibition constants were calculated using the Dixon method (44). RESULTS Cocrystal Structures of tACE-G13 with NoVel Inhibitors. The C domain-selective compounds kAW and kAF were derived from the moderately domain-selective ketone ACE inhibitor, kAP, by substitution of the P2′ group (Figure 1) (26, 29). To understand the mechanism of domain selectivity, we determined cocrystal structures of these two novel inhibitors with tACE, which is identical in sequence to the C domain. To facilitate crystallization, we used a minimally glycosylated construct, tACE-G13, containing only two intact glycosylation sites (28, 45). Structures were determined by molecular replacement and refined to final crystallographic R factors of 20.0% in both cases. Data processing and refinement statistics are presented in Table 1. In both the kAW and kAF cocrystal structures, the protein component is in a conformation identical to that of the inhibitor-free tACE-G13 structure (PDB entry 2IUL), as evidenced by all-atom root-mean-square (rms) deviations of 0.61 and 0.67 Å, respectively. Overall, this structure is mostly R-helical and ellipsoid in shape, containing two chloride ions, with the zinc ion buried deep in the active site cleft. As in the unliganded structure, glycan residues are present at the two intact glycosylation sites in tACE-G13, N72 and N109, where there is evidence of glycan-mediated stabilization of the lid helices, as observed previously (45). One inhibitor molecule is bound to the active site of each tACE-G13 molecule, in an extended conformation (Figure 2) similar to that seen for the phosphinic peptide inhibitor, RXPA380 (46). Weak density around the P2′ side chain of kAF indicates a degree of flexibility, which is also evidenced by the higher B factors of these atoms. These inhibitors have a ketone zinc-binding moiety analogous to the carbonyl oxygen of the scissile peptide bond; however, the density maps surprisingly indicate the stabilization of a hydrated geminal diolate (gem-diol) transition state at the active site, in which two oxygens are attached to the carbonyl carbon analogue (Figure 3). The zinc coordination distances for the gem-diol oxygens are 2.23 and 2.42 Å for kAW and 2.41 and 2.52 Å for kAF (see Table 1 of the Supporting Information). kAW makes nine polar contacts (potential hydrogen bonds and ionic interactions) with seven protein side chains, while

Watermeyer et al. Table 1: X-ray Data Collection and Processing Statistics resolution (Å)a no. of reflections redundancya completeness (%)a 〈I/σ(I)〉a no. of molecules per asymmetric unit space group a (Å) b (Å) c (Å) Rmerge (%)a,b no. of reflections in working set (test set) Rcryst (%)a,c Rfree (%)a,d mean B factor (Å2) (minimum-maximum) no. of solvent atoms no. of protein atomse no. of metal ions no. of inhibitor atoms no. of glycan atoms root-mean-square deviations from ideal stereochemistry bonds (Å) angles (deg) dihedrals (deg) improper angles (deg) Ramachandran plot, % residues in favored regions

kAW

kAF

43-2.18 (2.26-2.18) 130768 3.8 (3.2) 95.0 (95.2) 10.87 (2.37) 1

43-2.17 (2.25-2.17) 158385 4.8 (4.1) 88.3 (80.2) 14.36 (2.61) 1

P212121 59.85 85.04 134.92 10.9 (42.1) 31173 (1753)

P212121 60.38 85.07 135.51 8.7 (39.2) 30340 (1459)

20.0 (26.6) 23.9 (29.7) 21.6 (2.8-69.0)

20.0 (25.0) 24.3 (26.4) 22.7 (5.9-72.5)

254 4750 3 39 67

230 4777 3 36 67

0.01 1.4 22.2 1.2 97.40

0.01 1.3 21.2 1.0 96.23

a Values in parentheses refer to the highest-resolution shell. b Rmerge ) [∑h∑j|Ij(h) - 〈I(h)〉|]/∑h∑j|Ij(h)|, where Ii(h) and 〈I(h)〉 are the ith and mean measurements of the intensity of reflection h, respectively. c Rcryst ) ∑h|Fo - Fc|/∑hFo, where Fo and Fc are the observed and calculated structure factor amplitudes of reflection h of the working set, respectively. d Rfree is equal to Rcryst for h belonging to the test set of reflections. e Excluding duplicated atoms in alternate conformations.

kAF makes nine polar contacts with six protein side chains, the difference being small changes in the orientations of Q281 and E384 (Table 1 of the Supporting Information). Additional hydrogen bonds are made with four water molecules, the positions of which are conserved between the structures, as well as with a glycerol molecule derived from the cryoprotectant in the kAW structure. The conformation of the Trp moiety of kAW does not appear to be affected by the presence of the glycerol, since it takes the same conformation as the P2′ Trp of RXPA380, which does not have a glycerol in this position. Moreover, no glycerol-like density was present at this position in composite omit maps generated using the RXPA380 structure factors from the Protein Data Bank. These solvent molecules form part of a hydrogen bonding network connecting the inhibitor to other active site residues and to the bulk solvent (Figure 4). Close contacts are also observed between the nonpolar moieties of the inhibitors and 17 active site residues as well as a number of water molecules (11 and 13 for kAW and kAF, respectively), and glycerol and acetate ions in the case of kAW (Table 2 of the Supporting Information). Some of these active site residues (T282, S355, V379, V380, H387, F457, F512, V518, and F527) could contribute favorably to binding entropy by the hydrophobic effect; others (Q281, H353, A354, E384, D415, and the water molecules) are polar

Domain Selectivity of Ketone ACE Inhibitors

Biochemistry, Vol. 47, No. 22, 2008 5945

FIGURE 2: Stick representation of kAW (A) and kAF (B) in the active site, showing the first Fobs - Fcalc difference density map (purple mesh) into which the inhibitor model was built, contoured at a level of 3σ. The position of the active site zinc ion (ZN2) is shown as a green nonbonded sphere, indicated with an arrow, and zinc-binding residues are shown as black lines in the foreground. Clear density for a tetrahedral gem-diol hydrated form of the zinc-binding ketone moiety is evident. Inhibitor-contacting residues that differ between domains are colored yellow (tACE residues from the cocrystal structure) and purple (N domain residues from the aligned N domain structure; PDB entry 2C6F). Residue labels are given as tACE/N domain, and the P1, P2, and P2′ moieties are labeled. In panel A, the active site-associated glycerol (GOL) and acetate (ACT) molecules are also shown as yellow sticks in Fobs - Fcalc difference density.

FIGURE 3: Stabilization of the hydrated transition state in the active site. The ketone inhibitor kAW (blue sticks) in Fobs - Fcalc density (purple mesh, contoured at 3σ) has been oriented to show the gemdiol moiety coordinating the active zinc ion (gray sphere). The zinccoordinating residues and those involved in stabilizing the intermediate are labeled and shown as yellow sticks.

or charged groups, and some (H383 and Y523) make both favorable and unfavorable contacts.

Nine residues that interact with kAW and kAF directly, or indirectly via solvent atoms, are replaced with a different amino acid in the N domain and may thus contribute to the domain selectivity of these ketone inhibitors (Figure 2). Kinetic Analysis of ActiVe Site ACE Mutants. To investigate the role of the nine active site residues that differ between domains, a series of tACE mutants was generated in which these residues were converted to their N domain counterparts (Table 2). All of the mutants had Km values similar to that of the wild type, and while there was some variation in kcat values, none were significantly lower than that of the wild type (data not shown). Using the substrate Abz-FRK(Dnp)P-OH under the conditions described here, kAW shows strong C domain selectivity, having almost 1300-fold greater affinity for the C domain than for the N domain (Table 2). The greatest decreases in affinity for kAW observed among the mutants were with the S1 pocket V518T and S2 F391Y mutants (Table 2 and Figure 5). The mutants E403R and S516N in these pockets displayed

5946

Biochemistry, Vol. 47, No. 22, 2008

Watermeyer et al.

FIGURE 5: Comparison of the relative binding affinity of tACE active site mutants for ketone inhibitors kAW (black bars) and kAF (gray bars) with that of wild-type tACE (C domain). Values above the line represent an N domain-like decrease in affinity relative to that of tACE. The N domain Ki values were 1258- and >602-fold higher than that of tACE for kAW and kAF, respectively.

FIGURE 4: Close-up views of the S2′ pocket with kAW (A) and kAF (B), showing the solvent network. The inhibitors and glycerol are shown as yellow sticks in Fobs - Fcalc difference density (purple mesh), contoured at 3σ. The active site zinc ion is shown as a green nonbonded sphere, water molecules are shown as light blue nonbonded spheres, and solvent hydrogen bonds are shown as blue dashed lines. Table 2: Kinetics of Inhibition by kAW and kAP of tACE Mutants Having Active Site Residues Mutated to Their N Domain Counterpartsa pocket S2′ S2′ S2′ S2′ S2′ S2′ S1 S1 S2 S2

tACE mutant constructb

kAW Ki (µM)

kAF Ki (µM)

kAP Ki (µM)

tACE T282S(260) E376D(354) V379S(357) V380T(358) V379S/V380T D453E(431) S516N(494) V518T(496) F391Y(369) E403R(381) N domain

0.679 0.920 2.33 0.064 2.65 0.870 0.618 0.497 9.74 4.75 0.24 854.2

0.83 ndc 2.4 0.81 1.9 ndc ndc ndc 18.1 3.9 ndc >500

0.05 ndc ndc ndc ndc ndc ndc 0.012 1.3 0.43 ndc 1.5

a Inhibition constants were determined using the Dixon method (44) with fluorogenic peptide substrate Abz-FRK(Dnp)P-OH. b tACE residue and number, followed by the corresponding N domain residue to which it was mutated, with N domain numbering in parentheses. c Not determined.

no change in affinity for kAW relative to that of wild-type tACE. In the S2′ pocket, mutants E376D and V380T demonstrated similar increases in Ki, while the V379S mutation caused an increase in affinity for the inhibitor. The

T282S and D453E mutants displayed Ki values similar to that of the wild-type C domain (Table 2 and Figure 5). The selectivity of kAW is only 2-fold lower than that of RXPA380, which shares a P2′ Trp, P1 Phe, and P2 benzyl moiety with kAW (14). However, the affinity of kAW for the C domain is approximately 230-fold lower than that of RXPA380, which has a C domain Ki of 3 nM (14). These data for RXPA380 were determined under the same reaction conditions as those used here, but using a different fluorogenic peptide substrate, Mca-ASDK-DpaOH (14). No change in selectivity was observed when Abz-FRK(Dnp)P-OH was used (data not shown). kAF was also shown to be highly C domain-selective under these conditions, with a C domain Ki of 0.83 µM (Table 2). No inhibition of the N domain was observed using up to 500 µM kAF and at a range of chloride concentrations. Because of the limited solubility of this compound, inhibition could not be tested at higher concentrations. The V518T mutation again displayed the greatest decrease in affinity for kAF, while small decreases in affinity were also observed for the F391Y, E376D, and V380T mutants (Table 2 and Figure 5). To understand the domain selectivity of the parent compound, kAP, we tested the effects of the F391Y, S516N, and V518T mutations in the S1 and S2 pockets on the binding affinity of this inhibitor. kAP had a 30-fold higher affinity for the C domain than for the N domain (Table 2). Notably, the F391Y mutation in the S2 pocket caused an 8-fold decrease in affinity relative to the wild-type C domain, while the S1 V518T mutant had a Ki similar to that observed for the N domain. The S516N substitution resulted in a binding affinity comparable to that of the wild-type C domain (Table 2). DISCUSSION As an enzyme with two homologous catalytically active domains, ACE is an intriguing drug target. Differences in the activity profiles of the domains, particularly with reference to the cleavage of angiotensin I and bradykinin, render it desirable to design selective inhibitors to the C domain; however, the high degree of homology between the domains makes this a complex problem. That domain-selective inhibition is possible is evident from the selectivity of some known substrates and inhibitors, and from the recent identification of the strongly domain-selective phosphinic

Domain Selectivity of Ketone ACE Inhibitors peptide inhibitors RXPA380 and RXP407 (13, 14, 24, 47, 48). We have recently developed new C domain-selective inhibitors kAW and kAF, based on the moderately domainselective inhibitor kAP (26, 29). In zinc metalloproteases, peptide bond hydrolysis is thought to occur via a tetrahedral transition state which forms upon nucleophilic attack of the carbonyl carbon by an activated zinc-bound water molecule at the active site (49, 50). This is followed by protonation of the scissile amide nitrogen and subsequent peptide bond cleavage. We found that the replacement of the scissile peptide bond nitrogen with a carbon atom in these inhibitors (Figure 1) led to the trapping of the tetrahedral gem-diol transition state at the active site (Figure 3). The hydrated transition state has been observed previously in the structures of carboxypeptidase A (CPA) cocrystallized with ketomethylene inhibitors 5-benzamido-2-benzyl-4-oxopentanoic acid (BOP) and 5-amino(N-tert-butoxycarbonyl)2-benzyl-4-oxo-6-phenylhexanoic acid (51, 52; structures not available in the Protein Data Bank). In these structures, the hydrated form of the ketone was observed at the enzyme active site despite the fact that this form does not occur (80% Efficiency. Strategies 9, 3–4. Ehlers, M. R., Chen, Y. N., and Riordan, J. F. (1991) Purification and characterization of recombinant human testis angiotensinconverting enzyme expressed in Chinese hamster ovary cells. Protein Expression Purif. 2, 1–9. Friedland, J., and Silverstein, E. (1976) A sensitive fluorimetric assay for serum angiotensin-converting enzyme. Am. J. Clin. Pathol. 66, 416–424. Chayen, N. E. (1997) The role of oil in macromolecular crystallization. Structure 5, 1269–1274. Otwinowski, Z., and Minor, W. (1997) Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307–326. Kissinger, C. R., Gehlhaar, D. K., and Fogel, D. B. (1999) Rapid automated molecular replacement by evolutionary search. Acta Crystallogr. D55, 484–491. Jones, T. A. (1978) A graphics model building and refinement system for macromolecules. J. Appl. Crystallogr. 11, 268–272. Brunger, A. T., Adams, P. D., Clore, G. M., DeLano, W. L., Gros, P., Grosse-Kunstleve, R. W., Jiang, J. S., Kuszewski, J., Nilges, M., Pannu, N. S., Read, R. J., Rice, L. M., Simonson, T., and Warren, G. L. (1998) Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr. D54, 905–921. Schuttelkopf, A. W., and van Aalten, D. M. (2004) PRODRG: A tool for high-throughput crystallography of protein-ligand complexes. Acta Crystallogr. D60, 1355–1363. Collaborative Computational Project, No. 4. (1994) The CCP4 suite: Programs for protein crystallography. Acta Crystallogr. D50, 760– 763. Davis, I. W., Murray, L. W., Richardson, J. S., and Richardson, D. C. (2004) MOLPROBITY: structure validation and all-atom contact analysis for nucleic acids and their complexes. Nucleic Acids Res. 32, W615-W619. McDonald, I. K., and Thornton, J. M. (1994) Satisfying hydrogen bonding potential in proteins. J. Mol. Biol. 238, 777–793. Eisenthal, R., and Cornish-Bowden, A. (1974) The direct linear plot. A new graphical procedure for estimating enzyme kinetic parameters. Biochem. J. 139, 715–720. Carmona, A. K., Schwager, S. L., Juliano, M. A., Juliano, L., and Sturrock, E. D. (2006) A continuous fluorescence resonance energy transfer angiotensin I-converting enzyme assay. Nat. Protoc. 1, 1971–1976. Dixon, M. (1953) The determination of enzyme inhibitor constants. Biochem. J. 55, 170–171. Watermeyer, J. M., Sewell, B. T., Schwager, S. L., Natesh, R., Corradi, H. R., Acharya, K. R., and Sturrock, E. D. (2006) Structure of testis ACE glycosylation mutants and evidence for conserved domain movement. Biochemistry 45, 12654–12663. Corradi, H. R., Chitapi, I., Sewell, B. T., Georgiadis, D., Dive, V., Sturrock, E. D., and Acharya, K. R. (2007) The structure of testis angiotensin-converting enzyme in complex with the C domain-specific inhibitor RXPA380. Biochemistry 46, 5473–5478. Hayashi, M. A. F., and Camargo, A. C. M. (2005) The Bradykininpotentiating peptides from venom gland and brain of Bothrops jararaca contain highly site specific inhibitors of the somatic angiotensin-converting enzyme. Toxicon 45, 1163–1170. Georgiadis, D., Cuniasse, P., Cotton, J., Yiotakis, A., and Dive, V. (2004) Structural determinants of RXPA380, a potent and highly selective inhibitor of the angiotensin-converting enzyme C-domain. Biochemistry 43, 8048–8054.

5950

Biochemistry, Vol. 47, No. 22, 2008

49. Lipscomb, W. N., and Strater, N. (1996) Recent advances in zinc enzymology. Chem. ReV. 96, 2375–2433. 50. Pelmenschikov, V., Blomberg, M. R., and Siegbahn, P. E. (2002) A theoretical study of the mechanism for peptide hydrolysis by thermolysin. J. Biol. Inorg. Chem. 7, 284–298. 51. Christianson, D. W., David, P. R., and Lipscomb, W. N. (1987) Mechanism of Carboxypeptidase A: Hydration of a Ketonic Substrate Analogue. Proc. Natl. Acad. Sci. U.S.A. 84, 1512–1515. 52. Shoham, G., Christianson, D. W., and Oren, D. A. (1988) Complex between carboxypeptidase A and a hydrated ketomethylene substrate analogue. Proc. Natl. Acad. Sci. U.S.A. 85, 684–688. 53. Strater, N., and Lipscomb, W. N. (1995) Transition-State Analog L-Leucinephosphonic Acid Bound to Bovine Lens Leucine Aminopeptidase: X-Ray Structure at 1.65 Angstrom Resolution in a New Crystal Form. Biochemistry 34, 9200–9210. 54. Mangani, S., Carloni, P., and Orioli, P. (1992) Crystal structure of the complex between carboxypeptidase A and the biproduct analog inhibitor L-benzylsuccinate at 2.0 Å resolution. J. Mol. Biol. 223, 573–578. 55. Coates, L., Erskine, P. T., Mall, S., Williams, P. A., Gill, R. S., Wood, S. P., and Cooper, J. B. (2003) The structure of endothiapepsin complexed with the gem-diol inhibitor PD-135,040 at 1.37 angstrom. Acta Crystallogr. D59, 978–981.

Watermeyer et al. 56. Zhang, Y., Desharnais, J., Marsilje, T. H., Li, C., Hedrick, M. P., Gooljarsingh, L. T., Tavassoli, A., Benkovic, S. J., Olson, A. J., Boger, D. L., and Wilson, I. A. (2003) Rational design, synthesis, evaluation, and crystal structure of a potent inhibitor of human GAR Tfase: 10-(Trifluoroacetyl)-5,10-dideazaacyclic-5,6,7,8-tetrahydrofolic acid. Biochemistry 42, 6043–6056. 57. Natesh, R., Schwager, S. L., Sturrock, E. D., and Acharya, K. R. (2003) Crystal structure of the human angiotensin-converting enzyme-lisinopril complex. Nature 421, 551–554. 58. Natesh, R., Schwager, S. L., Evans, H. R., Sturrock, E. D., and Acharya, K. R. (2004) Structural details on the binding of antihypertensive drugs captopril and enalaprilat to human testicular angiotensin I-converting enzyme. Biochemistry 43, 8718–8724. 59. Michaud, A., Chauvet, M. T., and Corvol, P. (1999) N-Domain selectivity of angiotensin I-converting enzyme as assessed by structure-function studies of its highly selective substrate, N-acetylseryl-aspartyl-lysyl-proline. Biochem. Pharmacol. 57, 611–618. 60. Acharya, K. R., Sturrock, E. D., Riordan, J. F., and Ehlers, M. R. (2003) ACE revisited: A new target for structure-based drug design. Nat. ReV. Drug DiscoVery 2, 891–902. BI8002605